How are OCM jobs recreated by catbundle.sql?

  • From: Rich J <rjoralist3@xxxxxxxxxxxxxxxxxxxxx>
  • To: Oracle L <oracle-l@xxxxxxxxxxxxx>
  • Date: Wed, 14 Oct 2015 08:54:47 -0500



Hey all,

So, I realized that almost all of my 11.2.0.3.8 DBs (on AIX 7.1) say
they're 11.2.0.3.0, with no entries in DBA_REGISTRY_HISTORY. Doh.
Apparently catbundle.sql needs to run for new DBs created after PSU
installs. So I do that, on the dev DB server of course.

10PM last night, my log checker picks up the ignorable "ERROR:
kfnUseConn - failure to make a connection" error in scheduler job logs
for each DB. Checking, I see that the disabled ORACLE_OCM jobs
MGMT_CONFIG_JOB and MGMT_STATS_CONFIG_JOB have either been recreated or
reenabled. I disable those jobs as part of my DB create script since the
MOS sync is handled by EM12.

The kicker is that I can't figure out how the jobs were recreated. Prior
to running catbundle.sql, I went through its code. It doesn't call any
procs in ?/rdbms/admin and in rechecking, it doesn't appear to do
anything with OCM. I went through the files in ?/rdbms/admin and traced
the ocmjb10.sql (it creates the OCM jobs) back to everything that calls
it, but again there doesn't seem to be anything related to
catbundle.sql.

Thoughts anyone? This has me wondering what else I missed before I run
catbundle.sql in Production....

Thanks!
Rich

Other related posts: